On 12/09/2012 10:06 AM, Sebastian Gabler wrote:
Am 07.12.2012 20:16, schrieb Jörn Nettingsmeier:
as mentioned by someone else, rme hdsps can be ganged. _but_ i've seen
weird timing issues with an old hdsp madi pci and a hdsp madi pcie -
they would be in perfect sync, but with random offsets of plus/minus
several hundred samples, apparently randomly after each reboot.
chances are this bug has long since been solved, but do measure it and
complain to rme loudly if you still see it.
I have seen that with all HDSP(e) cards. This happens due to the RME
SteadyClock sync mechanism if the clock source is not steady. Provide a
clock source that is present before the computer boots, have set all RME
cards to external sync, and you should be fine. If the clock is lost
during operation (i.e. connection loss), it should suffice to restart
the audio server or ASIO loop to get back to stable ground.
Is this a bug? IMHO, no. I think it's quite logical that a system that
tries to mitigate clock drifts will do exactly that. But documentation
could help with the multi-card scenario. You cannot use internal clock
reference of one of the cards in this case because it will not be steady
during re-boot. Each of the cards has an independent quartz, and there
is no way to predict if the master will be "ready" before the slaves.
interesting comment, thanks. i'm pretty sure that the computer system as
a whole was slaved to a nexus clock which was definitely running all the
time, but i'd have to ask my colleague if he remembers how the cards
themselves were connected. i think we used a short wc cable from the
first card to the second, or maybe the internal sync cable.
i could imagine how a situation like you describe could come about when
wiring the cards this way, but it feels a bit weird. iiuc, connecting
each card to an always-active wc source directly should have avoided the
issue, right? when i get the chance, i'll check that.
--
Jörn Nettingsmeier
Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487
Meister für Veranstaltungstechnik (Bühne/Studio)
Tonmeister (VDT)
http://stackingdwarves.net
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
